Application logging is done via log4j into ./logs/OfferNet.log local file. There are three logging levels configured:
INFO: logging messages following the harmonized pattern which are important events to be sent to ElasticStack for processing and eventually producing reports / analysis of computational experiments;
WARN: messages that could be (or not) formatted to the above pattern, but which do not need to be sent to the ES;
DEBUG: freely formatted messages which will not be sent to ES and do not need to be archived / analyzed, but may be necessary for understanding bugs / internals;
Logging level can be changed in ./configs/log4j-properties.groovy; INFO is preferred as it minimized the amount of messages to be processed by local filebeat process and sent via the network for ES to process.
On ES side, this pull request includes the logstash pipeline written so that it parses log messages into key value pairs / fields. Since they end up in Elastic Search this allows them to be efficiently searched and analyzed in the future (and archived).
Application logging is done via log4j into
./logs/OfferNet.loglocal file. There are three logging levels configured:./configs/log4j-properties.groovy; INFO is preferred as it minimized the amount of messages to be processed by local filebeat process and sent via the network for ES to process.On ES side, this pull request includes the logstash pipeline written so that it parses log messages into key value pairs / fields. Since they end up in Elastic Search this allows them to be efficiently searched and analyzed in the future (and archived).
Relates to #37